Motion Compensation Search Constellation for Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current motion compensation methods in video coding, such as MMVD technology, face challenges in accurately expressing motion information due to limited search directions and step lengths, leading to insufficient coding efficiency and accuracy in representing motion conditions of most objects.
Innovation Solution
The proposed method expands search directions to include diagonal directions like upper left, lower left, upper right, and lower right, and allocates search step lengths differently in horizontal, vertical, and diagonal directions to cover a wider search range, constructing a search position constellation with an initial motion vector as the center point, allowing for more accurate motion vector determination and improved coding performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If search directions are limited to horizontal and vertical directions only, then device complexity is reduced, but measurement precision of motion information is insufficient
Solution Approach 1:
The patent extends the search from traditional horizontal and vertical directions to include diagonal directions, adding a new dimension to the search space. This allows the motion compensation to capture diagonal motion patterns that were previously missed, improving motion information accuracy without significantly increasing overall system complexity.
Solution Approach 2:
The search position constellation is segmented into different directional components (horizontal, vertical, and diagonal directions). Each direction has its own set of search points arranged according to specific rules, allowing independent optimization of each directional segment while maintaining overall system manageability.
2Manufacturing precision
If uniform step lengths are used in all directions, then device complexity is reduced, but manufacturing precision of motion vector expression is insufficient
Solution Approach 1:
Different step lengths are allocated to different directions based on local motion characteristics. Diagonal directions receive different step length treatments compared to horizontal and vertical directions, allowing each direction to be optimized for its specific motion patterns. This local optimization improves motion vector expression accuracy while keeping the allocation rules systematic and manageable.
3Measurement precision
If search range is limited to cover only cardinal directions, then productivity of encoding process is improved, but measurement precision of diagonal motion is insufficient
Solution Approach 1:
The patent adds diagonal search directions as an extension beyond the traditional cardinal directions. Rather than completely redesigning the search process, it partially extends the search coverage to include diagonal points, achieving improved diagonal motion detection while maintaining the efficient cardinal direction search framework.
Data Source
AI summary
Processing methods for motion compensation, an encoder, and a decoder are provided. The method includes the following. Determine a search position constellation with a point pointed by an initial motion vector as a center point, where the search position constellation includes N directions, search for at least one search point by using at least one type of step lengths in the directions, where the at least one search point at least includes at least one search point in diagonal directions. Search for at least one search position from the search position constellation based on a preset range coverage rule, where the preset range coverage rule is to arrange search points in the directions for search. Obtain a new motion vector according to the at least one search position, and perform motion compensation based on the new motion vector to obtain a prediction value of a coding unit (CU).